better-search
Science Score: 62.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
○DOI references
-
✓Academic publication links
Links to: arxiv.org -
○Academic email domains
-
✓Institutional organization owner
Organization emory-irlab has institutional domain (ir.mathcs.emory.edu) -
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (10.6%) to scientific vocabulary
Repository
Basic Info
- Host: GitHub
- Owner: emory-irlab
- License: apache-2.0
- Default Branch: main
- Homepage: https://arxiv.org/abs/2311.11226
- Size: 37.1 KB
Statistics
- Stars: 1
- Watchers: 4
- Forks: 1
- Open Issues: 0
- Releases: 0
Metadata Files
README.md
BetterSearch 🔎
This is the repository where the user interface of the BetterSearch system would be released soon.
Paper Link: An Interactive Query Generation Assistant using LLM-based Prompt Modification and User Feedback
Video: Better Search IARPA Demonstration
BetterSearch, a cross-lingual search interface that supports automatic and interactive query generation over the BETTER dataset. BetterSearch provides a simple document search interface that displays documents in their original language along with their English translations, making it simple for researchers to navigate and analyze search results. The tool also supports diverse query generation, allowing users to explore search results more comprehensively. More importantly, it combines search with a prompting-based query generation interface which permits users to refine their queries and prompts with retrieval information. The BetterSearch interface could work as an effective starting template for performing qualitative analysis over other information retrieval experiments and datasets as well as serve as a tool to incorporate retrieval feedback and Human-In-The-Loop (HITL) studies.
Our lab developed the BetterSearch User Interface which is made up of 3 subsystems. Each of them are described below. Check this video for a complete demonstration. - Cross-Lingual Event based Retrieval - Automatic Query Generation - Prompting Based Interactive Query Generation & Feedback Each of the systems have been described on the IRLab webpage in detail.
Code
We recommend running the colab code shared as a part of the Query Explorer repository - which is a general version of this project and provides seamless integration with HuggingFace (for generation) and PyTerrier (for retrieval).
Citation
bibtex
@misc{dhole2023interactive,
title={An Interactive Query Generation Assistant using LLM-based Prompt Modification and User Feedback},
author={Kaustubh D. Dhole and Ramraj Chandradevan and Eugene Agichtein},
year={2023},
eprint={2311.11226},
archivePrefix={arXiv},
primaryClass={cs.AI},
url={https://arxiv.org/abs/2311.11226},
note={BETTER IARPA Demonstration Day}
}
Kaustubh Dhole, Ramaraj Chandradevan, Eugene Agichtein (Emory IR Lab) and the JHU Team
The project is funded by the Intelligence Advanced Research Projects Activity (IARPA).
Owner
- Name: Emory Intelligent Information Access Lab (IR Lab)
- Login: emory-irlab
- Kind: organization
- Email: emory.irlab@gmail.com
- Website: http://ir.mathcs.emory.edu/
- Repositories: 24
- Profile: https://github.com/emory-irlab
Citation (CITATION.cff)
cff-version: 1.2.0
message: "If you use this software, please cite it as below."
title: "An Interactive Query Generation Assistant using LLM-based Prompt Modification and User Feedback"
authors:
- family-names: Dhole
given-names: Kaustubh D.
- family-names: Chandradevan
given-names: Ramraj
- family-names: Agichtein
given-names: Eugene
year: 2023
version: 1.0.0
repository-code: "https://github.com/emory-irlab/better-search"
abstract: "BetterSearch, a cross-lingual search interface that supports automatic and interactive query generation over the BETTER dataset. BetterSearch provides a simple document search interface that displays documents in their original language along with their English translations, making it simple for researchers to navigate and analyze search results. The tool also supports diverse query generation, allowing users to explore search results more comprehensively. More importantly, it combines search with a prompting-based query generation interface which permits users to refine their queries and prompts with retrieval information. The BetterSearch interface could work as an effective starting template for performing qualitative analysis over other information retrieval experiments and datasets as well as serve as a tool to incorporate retrieval feedback and Human-In-The-Loop (HITL) studies."
identifiers:
- type: arxiv
value: 2311.11226
keywords:
- "Interactive Query Generation"
- "LLM-based Prompt Modification"
- "User Feedback"
- "Artificial Intelligence"
license: "Apache-2.0"
references:
- title: "An Interactive Query Generation Assistant using LLM-based Prompt Modification and User Feedback"
authors:
- family-names: Dhole
given-names: Kaustubh D.
- family-names: Chandradevan
given-names: Ramraj
- family-names: Agichtein
given-names: Eugene
year: 2023
doi: "10.48550/arXiv.2311.11226"
url: "https://arxiv.org/abs/2311.11226"